Table 7 Comparison of proposed deep learning architectures with pre-trained models based on accuracy value.

From: Artificial intelligence based classification and prediction of medical imaging using a novel framework of inverted and self-attention deep neural network architecture

Dataset

Deep learning model

Proposed 94-layered deep inverted residual

Proposed 84-layered self attention

Alexnet

GoogleNet

ResNet50

Densenet201

Breast cancer

98.5

98.4

93.5

92.7

95.1

96.7

Kvasir

94.9

95.1

90.2

90.7

92.8

93.3

ISIC2018

92.6

93.9

89.1

88.6

90.1

91.5

Lung Cancer

93.0

95.0

90.8

91.7

92.0

92.6

Oral Cancer

98.5

98.4

93.1

92.6

94.6

95.1

  1. The bold values denote the significant results.
Back to article page